Random effects regression mixtures for analyzing infant habituation

Random effects regression mixture models are a way to classify longitudinal data (or trajectories) having possibly varying lengths. The mixture structure of the traditional random effects regression mixture model arises through the distribution of the random regression coefficients, which is assumed to be a mixture of multivariate normals. An extension of this standard model is presented that accounts for various levels of heterogeneity among the trajectories, depending on their assumed error structure. A standard likelihood ratio test is presented for testing this error structure assumption. Full details of an expectation-conditional maximization algorithm for maximum likelihood estimation are also presented. This model is used to analyze data from an infant habituation experiment, where it is desirable to assess whether infants comprise different populations in terms of their habituation time.     

Small Town in Mass Society Revisited1

Abstract The case study of a small New York town that dramatized the thesis that the secular expansion of macro forces—urbanization, industrialization, bureaucratization—has permanently reduced the autonomy of all small communities is an example of a special type of discovery/persuasion strategy in the social sciences: the “opposition case study.” In contrast to the more rigorous “competitive test” or the atheoretical “negative case,” opposition case studies confront the dominant perspective with a qualitative illustration of a new theory in the context of a zero-sum game. When they are successful, opposition cases meet four criteria: the dominant view is immediately rendered obsolete; the origin of the new idea supports its plausibility; the new perspective is shown to be testable; and the new perspective quickly generates new lines of research. Small Town in Mass Society meets the first criterion, and may have been heuristic, but its probable origin in populist ideology undermines its testability.     

IDENTITY DILEMMAS OF CHRONICALLY ILL MEN

Chronic illness frequently comes to men suddenly with immediate intensity, severity, and uncertainty. Because men contract more serious and life-threatening chronic illnesses than women, experiencing illness causes men different identity dilemmas. This paper explores men's identity dilemmas by studying how men experience chronic illnesses and by looking at how assumptions about masculinity affected their identity. The paper explores four major processes: (1) awakening to death after a lifethreatening crisis, (2) accommodating to uncertainty as men realize that the crisis has lasting consequences for their lives, (3) defining illness and disability and (4) preserving self to maintain a sense of coherence while experiencing loss and change. The data are derived from forty in-depth formal interviews of twenty men, informal interviews with these men, and an extensive collection of published and unpublished personal accounts. The data were analyzed through the strategies of grounded theory.     

The Impact of Question and Respondent Characteristics on Comprehension and Mapping Difficulties

Behavior coding is one technique researchers use to detect problemsin survey questions, but it has been primarily explored as apractical tool rather than a source of insight into the theoreticalunderstanding of the cognitive processes by which respondentsanswer survey questions. The latter is the focus of the currentinvestigation. Using data from a large study in which face-to-faceinterviews were taped and extensive behavior coding was done,we tested whether sets of respondent behavior codes could beused to distinguish respondent difficulties with comprehensionof the question from difficulties associated with mapping ajudgment onto the response format provided, and whether characteristicsof the survey questions and respondents could be used to predictwhen and for whom such difficulties would occur. Sets of behaviorcodes were identified that reflected comprehension and mappingdifficulties, and these two types of difficulties were associatedwith different question and respondent characteristics. Thisevidence suggests that behavior coding shows promise as a toolfor researchers studying the cognitive processes involved inanswering survey questions.     

Goal Ambiguity and Organizational Performance in U.S. Federal Agencies

In spite of numerous observations that government organizationshave high levels of organizational goal ambiguity that exertmajor influences on their other characteristics, few researchershave measured goal ambiguity and tested these frequent assertions.In previous research, we developed measures of four dimensionsof goal ambiguity: mission comprehension ambiguity, directivegoal ambiguity, evaluative goal ambiguity, and priority goalambiguity. Confirming hypotheses developed from the literatureon public organizations, the latter three variables showed relationsto such organizational characteristics as organizational age,financial publicness (proportion of funding from governmentallocations), and regulatory status. This article reports asecond analytical step of examining the relations between thegoal ambiguity dimensions and indicators of organizational performancebased on responses to the 2000 National Partnership for ReinventingGovernment Survey of federal employees. The performance variablesincluded managerial effectiveness, customer service orientation,productivity, and work quality. Regression analyses with numerouscontrol variables found that directive, evaluative, and prioritygoal ambiguity related negatively to managerial effectiveness.All four performance indicators showed significant negativerelationships with evaluative goal ambiguity and directive goalambiguity. The results provide further evidence of the viabilityof the new measures of goal ambiguity, support theory-basedbut previously untested hypotheses, and further indicate thefeasibility and value of analyzing goal ambiguity of governmentorganizations.     

The impact of unionization,right-to-work laws,and female labor force participation on earnings inequality across states

This study examines the effect of unionization, right-to-work laws, and participation of women in the labor force on income inequality. Two distinct models are developed using 1970 and 1980 census data on the 50 states in the U.S. First, an income inequality model is specified as a beta distribution of the second kind to estimate Gini measures of income inequality. Second, these Gini estimates are used in a simultaneous equations model. The 1970 results indicate that higher unionization rates decreased inequality while right-to-work laws increased inequality. In 1980, the measure of inequality was lower in states with higher female labor force participation. We thank an unknown referee and the editor for comments and criticisms that greatly improved the paper. The usual caveat applies.